Flow -> DataFlow

.travis.yml

@@ -10,5 +10,5 @@ notifications:
 # uncomment the following lines to override the default test script
 script:
  - if [[ -a .git/shallow ]]; then git fetch --unshallow; fi
- - julia -e 'Pkg.clone("https://github.com/MikeInnes/Flow.jl")'
+ - julia -e 'Pkg.clone("https://github.com/MikeInnes/DataFlow.jl")'
  - julia -e 'Pkg.clone(pwd()); Pkg.build("Flux"); Pkg.test("Flux"; coverage=true)'

pkg.yml

@@ -1,2 +1,2 @@
 require:
-  MikeInnes/Flow
+  MikeInnes/DataFlow

src/Flux.jl

@@ -1,7 +1,7 @@
 module Flux
 
-using MacroTools, Lazy, Flow, Juno
+using MacroTools, Lazy, DataFlow, Juno
-import Flow: graphm, syntax, prewalk!, prewalk, postwalk, iscyclic,
+import DataFlow: graphm, syntax, prewalk!, prewalk, postwalk, iscyclic,
   Constant, constant, isconstant, value, inputs, thread!, value, inputs,
   Split, Group, group
 import Juno: Tree, Row

src/backend/tensorflow/graph.jl

@@ -1,5 +1,5 @@
 import Base: @get!
-import Flow: Constant, postwalk, value, inputs, constant
+import DataFlow: Constant, postwalk, value, inputs, constant
 import TensorFlow: RawTensor
 
 cvalue(x) = x
@@ -26,7 +26,7 @@ graph(::Input, x) = x
 graph(p::MaxPool, x) =
   nn.max_pool(x, [1, p.size..., 1], [1, p.stride..., 1], "VALID")
 
-graph(::Flow.Group, xs...) = (xs...,)
+graph(::DataFlow.Group, xs...) = (xs...,)
 
 graph(params::Associative, c::Conv2D, x) =
   nn.conv2d(x, graph(params, c.filter), [1,c.stride...,1], "VALID")
@@ -55,7 +55,7 @@ end
 function graph(params::Associative, model, args...)
   g = Flux.graph(model)
   g == nothing && return graph(model, args...)
-  Flow.iscyclic(g) && error("This model has a cycle; try unrolling it first.")
+  DataFlow.iscyclic(g) && error("This model has a cycle; try unrolling it first.")
   graph(params, g, args...)
 end
 

src/backend/tensorflow/recurrent.jl

@@ -5,7 +5,7 @@ type SeqModel
   state::Any
 end
 
-cgroup(xs...) = Flow.group(map(constant, xs)...)
+cgroup(xs...) = DataFlow.group(map(constant, xs)...)
 
 function makesession(model::Flux.Unrolled)
   sess = Session(Graph())

src/backend/tensorflow/tensorflow.jl

@@ -1,6 +1,6 @@
 module TF
 
-using ..Flux, Flow, TensorFlow, Juno
+using ..Flux, DataFlow, TensorFlow, Juno
 import Flux: accuracy, spliceinputs, detuple
 
 export tf

src/compiler/code.jl

@@ -1,12 +1,12 @@
 # TODO: proper escaping
 
-import Flow: mapconst, cse
+import DataFlow: mapconst, cse
 
 export @net
 
 function process_func(ex, params = [])
   @capture(shortdef(ex), (args__,) -> body_)
-  body = @> body MacroTools.flatten liftloops!(params) graphm Flow.il
+  body = @> body MacroTools.flatten liftloops!(params) graphm DataFlow.il
   body = mapconst(x -> x in params ? :(self.$x) : x, body)
   return args, body
 end
@@ -48,7 +48,7 @@ end
 function build_backward(body, x, params = [])
   iscyclic(body) && return :(error("Can't run backward pass on a cyclic graph"))
   Δs = invert(body)
-  back = IVertex{Any}(Flow.Do())
+  back = IVertex{Any}(DataFlow.Do())
   for param in params
     haskey(Δs, :(self.$param)) || continue
     ex = Δs[:(self.$param)]
@@ -76,7 +76,7 @@ function process_type(ex)
     (self::$T)($(args...),) = $(build_forward(body, args))
     back!(self::$T, Δ, $(args...)) = $(build_backward(body, args[1], pnames))
     update!(self::$T, η) = $(map(p -> :(update!(self.$p, η)), pnames)...)
-    graph(self::$T) = $(Flow.constructor(makegraph(body, args)))
+    graph(self::$T) = $(DataFlow.constructor(makegraph(body, args)))
     nothing
   end |> esc
 end
@@ -84,7 +84,7 @@ end
 function process_anon(ex)
   args, body = process_func(ex)
   @assert length(args) == 1
-  :(Flux.Capacitor($(Flow.constructor(makegraph(body, args))))) |> esc
+  :(Flux.Capacitor($(DataFlow.constructor(makegraph(body, args))))) |> esc
 end
 
 macro net(ex)

src/compiler/diff.jl

@@ -8,7 +8,7 @@ symbolic[:+] = (Δ, args...) -> map(_->Δ, args)
 function ∇v(v::Vertex, Δ)
   haskey(symbolic, value(v)) && return symbolic[value(v)](Δ, inputs(v)...)
   Δ = vertex(:back!, constant(value(v)), constant(Δ), inputs(v)...)
-  map(i -> @flow(getindex($Δ, $i)), 1:Flow.nin(v))
+  map(i -> @flow(getindex($Δ, $i)), 1:DataFlow.nin(v))
 end
 
 function invert(v::IVertex, Δ = :Δ, out = d())

src/compiler/graph.jl

@@ -1,4 +1,4 @@
-# TODO: most (all?) of this could be in Flow
+# TODO: most (all?) of this could be in DataFlow
 
 immutable ModelInput end
 

src/compiler/loops.jl

@@ -8,7 +8,7 @@ end
 Delay(name) = Delay(name, nothing)
 
 function liftloops!(ex, params)
-  ex = Flow.normedges(ex)
+  ex = DataFlow.normedges(ex)
   hidden = intersect((b.args[1] for b in ex.args), params)
   edges = Dict(h => gensym("edge") for h in hidden)
   declared = Dict(h => false for h in hidden)